@recode

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt

rmoralesaramburu
Posts: 52
Joined: Mon Nov 18, 2013 9:09 am

@recode

Postby rmoralesaramburu » Mon Apr 14, 2014 10:20 am

Hello,

I would like to know what is wrong for the following code

series ddddd=@recode((dfs_h>dfs_h(-1)) and (dfs<dfs_l),1,0)

It doesn't work, does @recode works with "and", "or" ??

I am trying @recode because I can't use the IF inside a FOR loop.

Thanks

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13306
Joined: Tue Sep 16, 2008 5:38 pm

Re: @recode

Postby EViews Gareth » Mon Apr 14, 2014 10:36 am

Nothing wrong with it that I can see:

Code: Select all

create u 100
series dfs_h = @rnd
series dfs=@rnd
series dfs_l=@rnd

series ddddd=@recode((dfs_h>dfs_h(-1)) and (dfs<dfs_l),1,0)
Follow us on Twitter @IHSEViews

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Mon Jun 18, 2018 5:53 am

Hi there,

how do i use 2 svector with @recode command?
(related files as attached, please check.)
Actually, the purpose is that if "cusip_alpha" = "scomp_cusp_al", i would like to create a column called DS_s, which is base on "type", in group Scusp, is anyone have better solution for this?
Thank you so much!!!!
Attachments
900A_Act1_Cusp_test.csv
(66.38 KiB) Downloaded 9226 times
900a_cusip.prg
(1.59 KiB) Downloaded 9094 times
SP500_cusp_s.csv
(316.31 KiB) Downloaded 9141 times
Best,
SK :D

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: @recode

Postby EViews Matt » Mon Jun 18, 2018 8:43 am

Hello,

You could use something like,

Code: Select all

series DS_s = @recode(cusip_alpha = scomp_cusp_al, type, na)

and then add DS_s to group Scusp afterword.

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Tue Jun 19, 2018 12:23 am

Hi Matt,

Thank you for your advise. :)
After i tried, it still have some problem, you may see the picture from the attachment
For example,
R4(cusip_alpha) = R3206(Scomp_cusp_al) = 000360206, then
R4(DS_S) should be 544294(type)
But, as you can see from the attchment, R4(DS_s) shows "NA"

i'm not sure is the reason that my data contain string or @recode command can't not be used like this way...
please help, thanks a lot!!!!!!! :D


Code: Select all

%dir="F:\Thesis\Eviews\St11_900A_cusip"
%dir1="F:\Thesis\DataFltr_St3\1.Filter\900A_FltrRe\900A_FltrRe_mrg\cusip_alpha\csv"
%dir2="F:\Thesis\Eviews\St11_900A_cusip\test_cusp\csv"

wfcreate(wf={%country}wf) u 1 5000

import {%dir}\SP500_cusp_s.csv ftype=ascii rectype=crlf skip=0 fieldtype=delimited na="NA" delim=comma colhead=1 eoltype=pad badfield=NA @freq U 1 @smpl @all
import {%dir1}\9a_Act1_CuspRun.csv ftype=ascii rectype=crlf skip=0 fieldtype=delimited na="NA" delim=comma colhead=1 eoltype=pad badfield=NA @freq U 1 @smpl @all
'read {%dir}\SP500_cusp_s.csv x1 x2 x3 x4 x5 Scomp_cusp_al $
'import {%dir2}\900A_Act1_N.csv delim=comma @id cusip_alpha @destid Scomp_cusp_al

series DS_s = @recode(scomp_cusp_al = cusip_alpha, type, na)

group cuspN scomp_cusp_al s_gvkey syear s_comp_cusip s_comp_tic s_crsp_cusip s_crsp_tic s_crsp_name DS_s
Attachments
recode command.JPG
recode command.JPG (217.66 KiB) Viewed 37209 times
Best,
SK :D

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: @recode

Postby EViews Matt » Tue Jun 19, 2018 9:49 am

Ah, you want to test if each value in scomp_cusp_al matches ANY value in cusip_alpha, and copy the corresponding type value. @recode is not the way to do this. Basically, you need to manually go through every observation in scomp_cusp_al and search for a matching observation in cusip_alpha.

Code: Select all

series DS_s
for !i = 1 to @obsrange
   if scomp_cusp_al(!i) <> "" then
      for !j = 1 to cusip_alpha.@last
         if scomp_cusp_al(!i) = cusip_alpha(!j) then
            DS_s(!i) = type(!j)
            !j = @obsrange
         endif
      next
   endif
next

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Thu Jun 21, 2018 8:20 am

Hi Matt,

i have another question that i need to create new columns ds_s_F/ds_c_F base on the ds_s/c_{%cntry}({%cntry}=a1, a2, dd1, dd2, dd3, dd4, dd5)
you may see the attached.
If one of ds_s/c_{%cntry} have 'value', ds_s/c_F = 'value', otherwise, = '0'
i not sure @recode command could use in this kind of issue, or not?

thanks a lot!!!
Attachments
擷取.JPG
擷取.JPG (249.55 KiB) Viewed 37136 times
Best,
SK :D

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: @recode

Postby EViews Matt » Thu Jun 21, 2018 9:31 am

You want to examine values within each row, yes? For that @recode could work well. However, it's probably easier to use other EViews facilities. For example, you can add the series/columns you want to check to a group and then use @rfirst to identify the first non-NA value for each row of the group. Since your data appears to have only one unique non-NA value per row, identifying the first non-NA value matches what I believe you're requesting. Here's some sample code for ds_s_F,

Code: Select all

%names = @wcross("ds_s_", "a1 a2 " + @wcross("dd", "1 2 3 4 5"))
group tmp {%names}
ds_s_F = @nan(@rfirst(tmp), 0)

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Fri Jun 22, 2018 12:17 am

Hi Matt,

Thank you so so so much~~~ :D
Best,
SK :D

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Wed Jun 27, 2018 3:36 am

Hi there,

i would like to do downsize within 2 files as attached.
and, create a new file that when obs = ds_s_F, and keep the obs.
how do i do it?
thanks!!!
Attachments
02.JPG
ds
02.JPG (162.73 KiB) Viewed 37010 times
01.JPG
return code
01.JPG (215.51 KiB) Viewed 37010 times
Best,
SK :D

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: @recode

Postby EViews Matt » Wed Jun 27, 2018 9:17 am

It's unclear to me which observations you want to keep. Regardless, if you can create a dummy series indicating which observations you want to keep, you can then contract the workfile page using that series and save it as a new workfile.

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Wed Jun 27, 2018 9:33 am

Hi Matt,

i would like to check whether the code in the 2nd pic exist in the column DS_S_F in the 1st pic
if not exist then drop the whole column of the code
for example,
if B1(867270) not exist in the DS_S_F, then drop the column

At the end, it will have a new file with the code that DS_S_F exist

i'm not sure you understand my explanation or not...?
:?:
Best,
SK :D

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: @recode

Postby EViews Matt » Wed Jun 27, 2018 10:04 am

I understand now, you want to eliminate columns (series), not observations. Once you've imported both files, you can go through each series name of the second file (using @wlookup) and see whether it is present in DS_S_F (similar to the code several posts ago) within the first file. If not, then delete the series.

shineK
Posts: 24
Joined: Sat Jun 09, 2018 12:23 am
Location: New Taipei City,Taiwan
Contact:

Re: @recode

Postby shineK » Wed Jun 27, 2018 10:56 am

Hi Matt,

i just try to use @wlookup command, but im not quite sure the function.
i attached 2 files, please check. Thanks!!!
Attachments
Dret_Mret_test.xlsx
(35.03 MiB) Downloaded 9128 times
CUSIP_FF_run.xlsx
DS_S_F
(53.69 KiB) Downloaded 9180 times
Best,
SK :D

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: @recode

Postby EViews Matt » Wed Jun 27, 2018 1:01 pm

Here's an example that gets you part way there. I'd forgotten that EViews will prepend "num" to numeric series names during an import, but that's easy to work around.

Code: Select all

wfopen CUSIP_FF_run.xlsx range=CUSIP_FF colhead=1 na="#N/A" @smpl @all
wfopen Dret_Mret_test.xlsx range=ACT1_RET_F colhead=1 na="NA" @freq D5 @id @date(obs) @smpl @all

copy CUSIP_FF_RUN::Cusip_ff_run\ds_s_f ds_s_f
%all_series = @wlookup("num*")
for %series {%all_series}
   !id = @val(@mid(%series, 4))
   if @min(@abs(ds_s_f - !id)) <> 0 then
      delete %series
   endif
next
delete ds_s_f


Return to “Programming”

Who is online

Users browsing this forum: No registered users and 25 guests